Hepatitis C recurrence is not adversely affected by the use of donation after cardiac death liver allografts

Authors


Abstract

Many factors can worsen a recurrent hepatitis C virus (HCV) infection after liver transplantation (LT). We sought to determine whether the use of donation after cardiac death (DCD) livers affects HCV recurrence. From January 2000 to June 2008, 37 HCV patients underwent LT with DCD allografts. The outcomes and severity of HCV recurrence were analyzed along with those for 74 matched control patients with HCV who received donation after brain death (DBD) livers. The 2 groups had similar donor and recipient characteristics, immunosuppression regimens, rates of acute cellular rejection (ACR), and HCV profiles. DCD patients had a higher incidence of primary nonfunction (19% versus 3%, P = 0.006) and significantly higher peak aspartate aminotransferase levels in comparison with DBD subjects, suggesting a greater degree of ischemia/reperfusion injury. Although the survival rates were not significantly different, DCD recipients had lower 1- and 5-year patient survival rates (83% and 69% versus 84% and 78%, respectively, P = 0.75) and graft survival rates (70% and 61% versus 82% and 74%, respectively, P = 0.24). Three hundred fourteen protocol and clinically indicated liver biopsy procedures were performed within 6 years after transplantation, and mixed modeling analysis showed that fibrosis progression rates were similar for the 2 groups (0.6 fibrosis units/year according to the Ishak modified staging system). The rates of severe HCV recurrence (retransplantation or death due to recurrent hepatitis C and/or the development of stage 4/6 fibrosis or worse within 2 years) were similar [3 DCD patients (8%) versus 11 DBD patients (15%), P = 0.38], and cytomegalovirus infection (hazard ratio = 7.9, P = 0.002, 95% confidence interval = 2.1-28.9) and ACR (hazard ratio = 6.2, P = 0.002, 95% confidence interval = 2.0-19.7) were the only independent risk factors for severe recurrence. In summary, although there was a trend of poorer overall outcomes in DCD patients, the use of DCD livers did not appear to adversely affect HCV recurrence after LT. Liver Transpl 16:1288-1295, 2010. © 2010 AASLD.

Recurrent hepatitis C virus (HCV) infection after liver transplantation (LT) is nearly universal and leads to chronic hepatitis in most patients and progression to cirrhosis in 20% to 40% of patients within 5 years after transplantation.1 A number of factors have been associated with more aggressive or accelerated HCV recurrence; these include virus-related factors (HCV RNA titers before transplantation2 and genotype3), donor-related and transplant-related factors [the use of donors older than 45-60 years,4, 5 ischemia/reperfusion injury in the allograft,6, 7 and acute cellular rejection (ACR)8], host-related variables (gender9 and race10), immunosuppression-related factors (the use of induction therapy11, 12 and steroid bolus, maintenance, and rapid withdrawal),10, 13 and concurrent viral infections [cytomegalovirus (CMV)14 and human immunodeficiency virus15].

A critical issue in LT is the shortage of organs. In order to expand the donor pool, livers from extended criteria donors such as older donors and donation after cardiac death (DCD) donors have been increasingly utilized. The use of DCD livers is associated with a significantly higher risk of graft failure16, 17 and an increased incidence of biliary complications.18, 19 Nevertheless, as we and others have recently reported, DCD LT can lead to good outcomes and survival benefits.17, 20, 21 Because HCV cirrhosis is the leading indication for LT in the United States, we wanted to determine whether the use of DCD allografts affects HCV recurrence. Some have suggested that ischemia/reperfusion injury may predict poor outcomes in HCV patients,7 whereas others have reported that biliary complications after LT are associated with worse hepatitis C recurrence.22 DCD livers by nature are subject to a higher degree of ischemia/reperfusion injury than donation after brain death (DBD) livers and, as mentioned previously, have a significantly higher incidence of biliary complications. In the current study, we examined the impact of DCD livers on patient and graft survival in HCV-infected patients and compared the severity of their HCV recurrence to that of a matched group of patients who received DBD livers.

Abbreviations:

ACR, acute cellular rejection; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; CI, confidence interval; CMV, cytomegalovirus; DBD, donation after brain death; DCD, donation after cardiac death; HCV, hepatitis C virus; HR, hazard ratio; INR, international normalized ratio; LT, liver transplantation; MELD, Model for End-Stage Liver Disease; MMF, mycophenolate mofetil; PNF, primary nonfunction; SVR, sustained virological response.

PATIENTS AND METHODS

From January 2000 to June 2008, 1627 adult patients underwent LT at the Starzl Transplantation Institute (University of Pittsburgh Medical Center, Pittsburgh, PA). We conducted a retrospective review of 560 patients (34%) who were HCV+, and we identified 37 HCV+ patients who received DCD livers. The donor and recipient characteristics of these patients were compared to those of 74 control patients with HCV who received DBD livers. Each DCD patient was randomly matched to 2 DBD subjects with respect to the time of transplant, patient age, native or physiological Model for End-Stage Liver Disease (MELD) score at the time of transplant, donor age, and presence or absence of hepatocellular carcinoma. The rates of primary nonfunction (PNF), biliary complications, retransplantation, ACR, and CMV infection were compared between the 2 groups. The primary endpoint of the study was the development of severe HCV recurrence (which was defined as HCV-related graft failure leading to retransplantation or death and/or the development of Ishak stage 4/6 fibrosis within 2 years after transplantation). Secondary endpoints included patient and graft survival. The examined donor and patient characteristics included the recipient age, gender, MELD score, and body mass index (BMI); the donor age, HCV status, and BMI; and the cold ischemia time. HCV-related factors analyzed included virological parameters (HCV RNA levels and HCV genotype) and histological parameters (fibrosis stage) as well as the sustained virological response (SVR) rate of patients undergoing interferon-α/ribavirin therapy. The DCD procurement protocol was described in detail previously.17 This study was approved by the University of Pittsburgh institutional review board.

All patients received 1 g of intravenous methylprednisolone intraoperatively. Tacrolimus was initiated immediately after transplantation, and trough levels were maintained at approximately 8 to 10 ng/mL. Subjects who displayed tacrolimus toxicity (eg, neurotoxicity23) were converted to cyclosporine. Before 2002, recipients routinely received steroids (with an initial dose of 200 mg/day on postoperative day 1 that was tapered to 20 mg/day on day 6) and were weaned off them after 6 to 12 months. A few subjects were enrolled in anti-thymocyte globulin12 or alemtuzumab11 induction therapy studies. Since 2005, tacrolimus monotherapy with no induction therapy and steroid avoidance has been used, with mycophenolate mofetil (MMF) or sirolimus added to patients with renal impairment (ie, those on renal replacement therapy or with a creatinine level > 2.0 mg/dL) or tacrolimus neurotoxicity. Liver biopsy samples were obtained when clinically indicated (eg, a rise in liver function tests), and a protocol biopsy was performed 1 year after transplantation and every 1 to 3 years thereafter. HCV recurrence was graded and staged by a group of dedicated LT pathologists using the Ishak modified histological activity index and staging.24 Fibrosis scores at the baseline were based on the first biopsy sample taken less than 1 year after transplantation. Biopsy-proven ACR was treated with a steroid bolus with or without a steroid taper. CMV infections were monitored by CMV polymerase chain reaction, and preemptive treatment with intravenous ganciclovir (10 mg/kg/day) or oral valganciclovir (900 mg twice a day) that was adjusted for renal function was used. Treatment with interferon-α and ribavirin (400-1200 mg by mouth daily) for 48 weeks was initiated in patients with biopsy-proven HCV recurrence. Interferon alfa-2b (Intron A, Schering-Plough, Kenilworth, NJ; 1-3 ×106 U subcutaneously 3 times a week) was used before 2001, and peginterferon alfa-2a (Pegasys, Roche, Nutley, NJ; 90-180 μg/week subcutaneously) or peginterferon alfa-2b (PegIntron, Schering-Plough; 1.0 μg/kg/week subcutaneously) was used afterwards. Quantitative HCV RNA was obtained every 3 to 6 months. Virological response was defined as clearance of HCV RNA from the serum, and patients with a virological response longer than 6 months after the cessation of therapy were regarded as having an SVR.

Statistical Analysis

Data were described with estimates of the central tendency (means and medians) and spread (standard deviations and ranges) for continuous data and with frequencies and percentages for categorical data. Baseline characteristics were compared with the Student t test, Wilcoxon rank-sum test, or chi-square test. Patient and graft survival and severe HCV recurrence were analyzed with Cox proportional hazards modeling. Time-dependent covariates were used when they were appropriate. The log-rank test was used to examine differences in survival rates, and logistic regression was used to measure risk factors for the development of severe recurrent hepatitis C. Mixed modeling25 was used to analyze the changes in fibrosis from the baseline to follow-up between study groups. This modeling technique was used for 3 reasons. First, this type of model makes inferences by using information on all available data from the entire cohort collected at all follow-up time points. Second, these models missing data more efficiently than other methods of analysis or data imputation. Third, mixed models are powerful tools for analyzing correlated data. All univariate models achieving a P value < 0.10 were entered into a multivariate model. Nonnormally distributed data were transposed when this was appropriate. P values < 0.05 were considered significant. All analyses were performed with SAS 9.2 (SAS Institute, Inc., Cary, NC) or SPSS 16.0 (SPSS, Inc., Chicago, IL).

RESULTS

The HCV+ DCD and DBD groups were well matched with respect to the recipient characteristics (age, gender, BMI, and MELD score; Table 1) and follow-up (37.1 ± 29.0 versus 38.3 ± 28.2 months, respectively). The donor characteristics (age, gender, and BMI), male donor to female recipient pairing, and cold ischemia times were also similar between the 2 groups. The total bilirubin, aspartate aminotransferase (AST), and alanine aminotransferase (ALT) levels and the international normalized ratios (INRs) 1 week after transplantation were comparable between the 2 cohorts; however, DCD patients had significantly higher peak AST levels in comparison with DBD subjects (P = 0.02) and also tended to have higher peak ALT levels (P = 0.06); suggesting a greater degree of ischemia/reperfusion injury in the DCD grafts. The mean DCD donor warm ischemia time (from the extubation of the donor to the perfusion of organs) was 18.9 ± 7.3 minutes. DCD patients had a significantly higher incidence of PNF versus DBD subjects [7 patients (19%) versus 2 patients (3%), respectively, P = 0.006]; all these patients underwent retransplantation. The incidence of biliary complications was higher in the DCD group, although the difference was not statistically significant [9 DCD patients (24%) versus 11 DBD patients (15%), P = 0.29]. Biliary complications in DCD subjects included bile leaks (2 patients), anastomotic strictures (6 patients), and ischemic cholangiopathy with or without bile casts (5 patients); some patients had more than 1 type of biliary problem. All were successfully managed with endoscopic retrograde cholangiopancreatography and stenting, except for 1 patient with ischemic cholangiopathy who required retransplantation. All 11 DBD patients had anastomotic strictures, except for 1 patient who developed stones and sludge in the common bile duct that were associated with a clogged T-tube, and none required retransplantation for biliary complications. There was a trend of poorer graft and patient outcomes in DCD patients (1- and 5-year graft survival rates of 70% and 61% for DCD patients and 82% and 74% for DBD patients, P = 0.24, and 1- and 5-year patient survival rates of 83% and 69% for DCD patients and 84% and 78% for DBD patients, P = 0.75; Fig. 1), although the differences were not statistically significant.

Table 1. Demographic Data for the HCV+ Patients Who Received DCD or DBD Liver Allografts
CharacteristicDCD (n = 37)DBD (n = 74)P Value
  1. NOTE: The data are presented as the number of patients and frequency (%) or as the mean and standard deviation.

Recipient age (years)51 ± 6.751 ± 6.20.6
Recipient gender (male:female)31 (84%):6 (16%)58 (78%):16 (22%)0.6
Recipient BMI (kg/m2)29.2 ± 4.928.7 ± 5.00.5
MELD score16.1 ± 7.316.2 ± 7.10.9
Donor age (years)37.9 ± 16.438.1 ± 16.20.9
Donor gender (male:female)27 (73%):10 (27%)43 (58%):31 (42%)0.1
Donor BMI (kg/m2)26.6 ± 5.328.3 ± 9.60.8
Male donors to female transplant recipients6 (16%)6 (8%)0.2
Cold ischemia time (minutes)672 ± 144680 ± 1620.9
Total bilirubin at 1 week (mg/dL)5.8 ± 4.15.4 ± 4.70.1
Total bilirubin at 1 month (mg/dL)3.0 ± 5.52.4 ± 3.00.4
ALT at 1 week (IU/L)139 ± 86128 ± 1000.2
AST at 1 week (IU/L)65 ± 5664 ± 870.3
INR at 1 week1.3 ± 0.31.2 ± 0.20.3
Peak ALT (IU/L)1462 ± 9721270 ± 15520.06
Peak AST (IU/L)4491 ± 31052960 ± 30600.02
ACR11 (30%)30 (40%)0.3
CMV infection8 (22%)21 (28%)0.5
Figure 1.

Patient and graft survival of HCV+ recipients of DCD or DBD LT.

In agreement with our previous finding that the majority of graft failures in DCD patients occurred early after transplantation,17 11 of the 14 DCD allograft failures (79%) in this series occurred within the first year of transplant. PNF was the predominant cause of graft failure and occurred in 7 patients. Two subjects died of sepsis and multiorgan failure (1 patient had a bile leak), another died from unknown causes, and 1 patient expired from graft failure secondary to HCV recurrence. Among the remaining 3 patients whose grafts failed after the first year, HCV recurrence occurred in 2 subjects, whereas the third patient died from a myocardial infarction. In the DBD group, 13 of the 20 graft failures (65%) occurred in the first year for the following reasons: operative death (1), PNF (2), HCV recurrence (3), cardiovascular causes (2), fungal sepsis (2), recurrent HCC (1), and respiratory failure (2). Late graft failure was due to recurrent hepatitis C in 4 patients, and multiorgan failure, de novo malignancy, and unknown causes were responsible in the remaining 3 subjects. There was a trend of higher retransplantation rates in DCD patients versus DBD patients [8 patients (22%) versus 6 patients (8%), P = 0.07], although this was not statistically significant. PNF was the predominant indication for retransplantation (7 patients), and it was followed by ischemic cholangiopathy (1 subject) in the DCD cohort, whereas PNF (2 patients), recurrent hepatitis C (3 patients), and the development of de novo malignancy in the allograft (1 patient) were the reasons for retransplantation in the DBD group.

The rates of biopsy-proven ACR [11 DCD patients (30%) versus 30 DBD patients (40%), P = 0.3] and the numbers of rejection episodes per patient (1.45 ± 0.7 episodes for DCD patients versus 1.43 ± 0.7 episodes for DBD patients, P = 0.9) were similar. All the patients received tacrolimus except for a minority of patients who were converted to cyclosporine [1 DCD patient (3%) versus 3 DBD patients (4%), P = 1.0]. MMF was used frequently in DCD patients (21, 57%) and DBD patients (46, 62%; P = 0.68), whereas sirolimus was used less [7 DCD subjects (19%) versus 11 DBD subjects (15%), P = 0.59]. The numbers of DCD and DBD patients who received steroids were comparable [17 (46%) versus 43 (58%), P = 0.23], and the median durations of prednisone use were also similar (7.7 months for DCD patients versus 6.8 months for DBD patients). The numbers of DCD patients (4, 10.8%) and DBD patients (5, 6.8%) who received induction therapy with either anti-thymocyte globulin or alemtuzumab were similar (P = 0.48). There was no significant difference in the rates of CMV infection in the 2 groups (Table 1).

HCV Recurrence, Graft Loss, and Fibrosis Progression

We compared HCV characteristics as well as the severity of HCV recurrence in DCD and DBD patients. There were no HCV+ DCD donors, whereas 16 DBD donors (22%) were HCV+ (P = 0.001).

The DBD control patients were a representative sample of the entire cohort of HCV+ patients who underwent transplantation at our center during the same time period; 95 of the 478 patients (20%) for whom the HCV donor status was available received HCV+ allografts. HCV genotyping data were available for 26 DCD recipients and 51 DBD recipients; 22 DCD patients (85%) and 42 DBD subjects (82%) had genotype 1 (P = 0.5). HCV RNA titers were available for 21 DCD patients (56.8%) and 47 DBD subjects (63.6%) 6 months after transplantation and for 25 DCD patients (67.6%) and 50 DBD patients (67.7%) 1 year after transplantation. There was no significant difference in the mean HCV RNA titers between the DCD and DBD groups 6 months after transplantation (6,400,675 ± 9,798,360 versus 10,782,306 ± 10,675,000 IU/mL, respectively, P = 0.4) or 12 months after transplantation (3,378,282 ± 7,418,599 versus 4,730,110 ± 8,244,737 IU/mL, respectively, P = 0.7). Half of the patients in each group received interferon-α and ribavirin for recurrent HCV [19 DCD patients (51%) versus 37 DBD patients (50%), P = 1.0], and the SVR rates were similar between the 2 cohorts [3 of 19 DCD patients (16%) versus 4 of 37 DBD patients (11%), P = 0.75].

The rates of severe HCV recurrence (HCV-related graft failure leading to retransplantation or death and/or the development of stage 4/6 fibrosis within 2 years after transplantation) were similar between the 2 groups [3 DCD patients (8%) versus 11 DBD patients (15%), P = 0.38], and there was no significant difference in the cumulative probability of developing severe HCV recurrence post-LT (Fig. 2). Graft loss due to recurrent hepatitis C occurred in 2 DCD patients (5%) and 7 DBD patients (9%, P = 0.6). Both DCD patients died from liver failure, whereas 4 DBD patients died from liver failure, and 3 underwent retransplantation. Five patients rapidly developed stage 4 fibrosis within 1 to 2 years after transplantation (1 DCD subject and 4 DBD subjects). With follow-up times ranging from 30.3 to 93.6 months from the time of transplant, 3 of these patients (including 1 patient admitting to alcohol use) developed portal hypertension with signs of decompensation. A univariate analysis of the entire cohort for donor-related, recipient-related, and virus-related risk factors (Table 2) demonstrated that male donor to female recipient pairing (P = 0.003), total bilirubin levels at 1 week (P = 0.008) and 1 month after transplantation (P = 0.1), peak AST levels (P = 0.04), donor HCV+ status (P = 0.002), CMV infection (P = 0.002), and ACR (P = 0.002) were all associated with the development of severe recurrent hepatitis C. Multivariate analysis revealed that ACR [hazard ratio (HR) = 6.2, P = 0.002, 95% confidence interval (CI) = 2.0-19.7] and CMV infection (HR = 7.9, P = 0.002, 95% CI = 2.1-28.9) were independent risk factors for severe HCV recurrence (Table 2).

Figure 2.

Cumulative incidence of severe HCV recurrence after LT in DCD and DBD patients.

Table 2. Univariate and Multivariate Whole Cohort Analyses Assessing the Associations of Donor and Recipient Covariates With Severe HCV Recurrence (111 HCV+ Patients)
Univariate Analysis
VariableP Value
DCD versus DBD0.5
Recipient age (years)0.5
Recipient BMI (kg/m2)0.5
MELD score0.9
Male donors to female recipients0.003
Donor BMI (kg/m2)1.0
Donor HCV status0.002
CMV infection0.002
Occurrence of biliary complications0.1
ACR0.002
Total bilirubin at 1 week (mg/dL)0.008
Total bilirubin at 1 month (mg/dL)0.1
Peak AST (IU/L)0.04
Multivariate Analysis
VariableHR (95% CI)P Value
ACR6.2 (2.0-19.7)0.002
CMV infection7.9 (2.1-28.9)0.002

We also compared the rates of fibrosis progression in DCD and DBD patients who underwent serial liver biopsy (protocol or clinically indicated) 1 year after LT or later. In the DCD group, 12 of the 37 subjects (32%) did not undergo biopsy more than 1 year after transplantation: 7 died in the first year (1 from HCV recurrence), 4 underwent retransplantation for PNF, and 1 patient underwent biopsy only once 5 months after transplantation. In the DBD cohort, 23 of 74 subjects (31%) did not undergo biopsy after 1 year: 13 died in the first year (3 from HCV recurrence), 2 patients underwent retransplantation for PNF, and 8 did not undergo biopsy beyond the first year after transplant. Two-thirds of the entire cohort [25 DCD subjects (67%) and 50 DBD subjects (68%)], therefore, had serial biopsy samples that could be evaluated for fibrosis progression. A total of 314 biopsies were performed in these 75 patients: 114 times in DCD patients (mean number of biopsy procedures per patient = 4.56) and 200 times in DBD subjects (mean number of biopsy procedures per patient = 3.92). The mean times from LT to the initial biopsy procedure in the DCD and DBD cohorts were 10.4 ± 12.7 and 9.1 ± 11.8 weeks, respectively, and the mean fibrosis scores at baseline were 0.9 and 0.81, respectively. Within-group analysis showed that both cohorts progressed over time (rate = 0.58 fibrosis units/year, 95% CI = 0.42-0.74, P = 0.001). Although the mean fibrosis stage at each year after transplantation was higher for DCD patients versus DBD patients (Fig. 3A), the differences were not statistically significant, and no difference in fibrosis progression rates was found between the groups with mixed modeling analysis (Fig. 3B).

Figure 3.

(A) Mean fibrosis stages of DCD and DBD patients up to 5 years after transplantation and (B) fibrosis progression rates measured by mixed modeling analysis of recipients of DCD or DBD LT.

DISCUSSION

DCD livers, when used appropriately, increase the donor pool, and our recent report on the outcomes of a large cohort of DCD recipients showed a benefit from the use of these organs in critically ill patients.17 Most studies have shown that DCD liver allografts lead to inferior outcomes in comparison with DBD livers (eg, lower graft survival rates and higher rates of PNF and biliary complications),16-18 but some have shown similar results between the cohorts.20, 21 There has been only one study examining the outcomes of DCD LT in HCV-infected recipients. Yagci et al.26 reported on 14 HCV+ patients who received DCD livers and found lower patient and graft survival in comparison with HCV+ subjects who received DBD allografts, although the outcomes did not appear to be related to early HCV recurrence. However, no in-depth analysis of donor and recipient factors and HCV recurrence was conducted by the authors. In the current study, patient and graft survival rates for DCD patients were lower than those for a well-matched cohort of DBD subjects (Fig. 1), although the differences were not statistically significant, probably because of the relatively small number of patients. DCD patients had a higher incidence of PNF (19%) in comparison with DBD subjects, and this PNF rate was high in comparison with the rates of other series,18, 19 most likely because of our longer cold ischemia times (these patients were part of a larger cohort of 141 DCD patients who had an overall PNF rate of 12%).17 Ischemia/reperfusion or preservation injury after LT has been linked to poorer outcomes in HCV patients.6, 22 Recipients of DCD livers had significantly higher peak AST levels in the first week after LT (Table 1), and this suggested a greater degree of early ischemia/reperfusion injury in comparison with DBD patients. Clinically, ischemia/reperfusion injury increases the risk for ACR and biliary complications.27, 28 Rejection episodes, along with the use of older donors, are perhaps the most significant risk factors for more severe HCV recurrence after LT, and this has suggested to some that ischemia/reperfusion may be a surrogate marker for these factors.7, 29 The occurrence of biliary complications has also been linked to worse hepatitis C recurrence,23 and our analysis showed that the DCD recipients had a higher incidence of biliary complications in comparison with the DBD patients (24% versus 15%, respectively), although the difference was not statistically significant. The association between ischemia/reperfusion injury and biliary complications with DCD livers prompted us to determine whether utilization of DCD allografts affected HCV recurrence after LT.

Taking into consideration the retrospective nature of this study and the relatively small number of patients, we found that although patient and graft survival rates were lower in DCD patients versus DBD recipients, the use of DCD livers did not appear to adversely affect hepatitis C recurrence after LT. DCD and DBD patients were well matched with respect to recipient and donor factors (Table 1). HCV characteristics (ie, the genotypes, mean HCV RNA titers at 6 and 12 months after transplantation, numbers of patients who received HCV treatment, and SVR rates) were also similar between the 2 cohorts, except for HCV+ donor livers, which were not used in DCD patients but were used in 22% of DBD patients. Twenty percent of the HCV patients who underwent transplantation at our center during the same period received HCV+ liver allografts, and this indicated that the DBD control patients were a representative patient sample of the entire cohort of HCV patients. The patient and graft survival rates of the DBD control patients were similar to those of the larger cohort of HCV patients who received DBD allografts (data not shown). HCV+ DCD organs are common and are being increasingly offered by organ procurement organizations. We are not averse to using these livers; however, poor overall donor quality (eg, advanced fibrosis in the biopsy sample, a long donor warm ischemia time, and/or the presence of other poor donor characteristics) has thus far precluded us from using any HCV+ DCD allografts. The 2 groups had similar patterns of calcineurin inhibitors, MMF, and steroid use, and none of the patients were rapidly weaned from maintenance steroids; this practice has been suggested to lead to more severe hepatitis C recurrence.30 The cumulative probabilities of developing severe hepatitis C recurrence (HCV-related graft failure and/or the development of stage 4 or greater fibrosis within 2 years) were similar for the 2 groups (Fig. 2). Univariate analysis demonstrated that male donor to female recipient pairing, total bilirubin levels at 1 week and 1 month after transplantation, peak AST levels, donor HCV+ status, CMV infection, and ACR were associated with severe HCV recurrence (Table 2); however, multivariate analysis showed that ACR (HR = 6.2, P = 0.002, 95% CI = 2.0-19.7) and CMV infection (HR = 7.9, P = 0.002, 95% CI = 2.1-28.9) were the only independent risk factors for severe recurrent HCV infection (Table 2). We also performed an in-depth analysis of 487 HCV patients who underwent primary LT in the same study period. Cox proportional hazards modeling showed that only ACR, CMV, and donor age were independent risk factors for severe HCV recurrence, and neither donor HCV status nor DCD livers were found to be predictive factors (data not shown). ACR is one of the most important risk factors significantly increasing the severity of recurrent hepatitis C, most likely because of steroid boluses and subsequent increases in immunosuppression.10, 31 Steroid boluses have been associated with elevations in serum HCV RNA levels of 4- to 100-fold.32 The occurrence of rejection leads to earlier recurrence of hepatitis C and has been linked to the development of cirrhosis in HCV patients.31, 33 In addition, the risk of death has also been shown to increase 3-fold in hepatitis C patients who experience an episode of rejection and 5-fold in those who develop steroid-resistant rejection.8 DCD and DBD patients had similar rates of ACR, so at least in this study, it appears that a greater degree of ischemia/reperfusion did not predispose DCD patients to a higher incidence of rejection, as suggested by some.28 The occurrence of CMV infection has also been associated with more severe HCV recurrence.14, 34 CMV has been purported to have an immunomodulatory effect on transplant recipients and to enhance overall immunosuppression; this may favor recurrent hepatitis C.14, 34 The progression of fibrosis was not different between the DCD and DBD recipients (Fig. 3). Although our analysis was somewhat limited because one-third of the patients did not undergo biopsies beyond the first year of transplant (because of early deaths or retransplants for the majority), each of the remaining 75 patients (25 DCD and 50 DBD subjects) averaged 4.2 biopsies per patient and had a total of 314 biopsies performed up to 6 years posttransplantation. DCD and DBD patients had similar baseline fibrosis stages, and the 2 cohorts had similar fibrosis progression rates (0.6 fibrosis units/year according to the Ishak modified staging system; Fig. 3B) as measured by mixed modeling analysis,25 a powerful modeling technique that can handle missing data more efficiently than other methods of analysis or data imputation. Other studies have shown that increasing donor age, female recipient gender, steroid boluses, and HCV genotypes 1 and 4 are covariates that increase HCV recurrence and the development of fibrosis.4, 10, 31

In conclusion, DCD HCV+ patients had lower graft and patient survival rates in comparison with DBD recipients, although the differences did not reach statistical significance, most likely because of the small number of patients. The majority of graft failures occurred in the first year after transplantation, and in this series, PNF was the predominant cause of graft loss in DCD patients. Despite its limitations, this study, controlling for host-related, donor-related, viral-related, and immunosuppression-related factors, has demonstrated that recipients of DCD livers appear to have HCV-related outcomes (fibrosis progression and development of severe HCV recurrence) similar to those of recipients of DBD livers. Decreasing the incidence of PNF and early graft loss will surely improve outcomes for DCD liver recipients. However, additional studies with more patients are required to fully determine how much hepatitis C patients can truly benefit from the use of DCD livers.

Ancillary